one noise variable, logistic regression
## [1] "*************************************************************"
## [1] "one noise variable, logistic regression"
## [1] "bSigmaBest 32"
## [1] "naive effects model"
## [1] "one noise variable, logistic regression naive effects model fit model:"
##
## Call:
## glm(formula = formulaL, family = binomial, data = trainData)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.8068 -1.0493 0.5770 0.9415 2.5190
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.18447 0.05074 3.635 0.000277 ***
## n1 2.20269 0.13545 16.262 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2772.6 on 1999 degrees of freedom
## Residual deviance: 2256.7 on 1998 degrees of freedom
## AIC: 2260.7
##
## Number of Fisher Scoring iterations: 6
##
## [1] "one noise variable, logistic regression naive effects model train mean deviance 1.62786601580457"
## [1] "one noise variable, logistic regression naive effects model test mean deviance 3.71500787962648"
## [1] "effects model, sigma= 32"
## [1] "one noise variable, logistic regression effects model, sigma= 32 fit model:"
##
## Call:
## glm(formula = formulaL, family = binomial, data = trainData)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.214 -1.204 1.141 1.151 1.329
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.06419 0.04941 1.299 0.19387
## n1 0.03702 0.01249 2.964 0.00304 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2772.6 on 1999 degrees of freedom
## Residual deviance: 2763.7 on 1998 degrees of freedom
## AIC: 2767.7
##
## Number of Fisher Scoring iterations: 3
##
## [1] "one noise variable, logistic regression Noised 32 train mean deviance 1.99361582203067"
## [1] "one noise variable, logistic regression Noised 32 test mean deviance 2.00456032886049"
## [1] "effects model, jacknifed"
## [1] "one noise variable, logistic regression effects model, jackknifed fit model:"
##
## Call:
## glm(formula = formulaL, family = binomial, data = trainData)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.3619 -1.1570 0.9662 1.1980 1.2169
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.04838 0.04731 -1.023 0.30650
## n1 -0.06366 0.01954 -3.258 0.00112 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2772.6 on 1999 degrees of freedom
## Residual deviance: 2761.8 on 1998 degrees of freedom
## AIC: 2765.8
##
## Number of Fisher Scoring iterations: 4
##
## [1] "one noise variable, logistic regression jackknifed train mean deviance 1.99219567357296"
## [1] "one noise variable, logistic regression jackknifed test mean deviance 2.00542702505421"
## [1] "********"
## [1] "one noise variable, logistic regression AverageManyNoisedModels"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.999 2.000 2.001 2.001 2.001 2.005
## [1] 0.001104472
## [1] "********"
## [1] "********"
## [1] "one noise variable, logistic regression JackknifeModel"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.999 2.001 2.003 2.004 2.005 2.023
## [1] 0.003992458
## [1] "********"
## [1] "********"
## [1] "one noise variable, logistic regression NaiveModel"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 3.522 3.872 4.044 4.041 4.204 4.570
## [1] 0.2316713
## [1] "********"
## [1] "********"
## [1] "one noise variable, logistic regression NoisedModel"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.999 2.001 2.002 2.002 2.003 2.010
## [1] 0.002048446
## [1] "********"
## [1] "********"
## [1] "one noise variable, logistic regression ObliviousModel"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.999 2.000 2.000 2.001 2.001 2.006
## [1] 0.00106767
## [1] "********"
## [1] "*************************************************************"
one variable, logistic regression
## [1] "*************************************************************"
## [1] "one variable, logistic regression"
## [1] "bSigmaBest 4"
## [1] "naive effects model"
## [1] "one variable, logistic regression naive effects model fit model:"
##
## Call:
## glm(formula = formulaL, family = binomial, data = trainData)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.1243 -1.1809 0.4704 1.1554 1.5778
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.4731 0.0542 8.73 <2e-16 ***
## x1 3.1777 0.2114 15.03 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2747.0 on 1999 degrees of freedom
## Residual deviance: 2434.7 on 1998 degrees of freedom
## AIC: 2438.7
##
## Number of Fisher Scoring iterations: 4
##
## [1] "one variable, logistic regression naive effects model train mean deviance 1.75629049009229"
## [1] "one variable, logistic regression naive effects model test mean deviance 1.74484448505444"
## [1] "effects model, sigma= 4"
## [1] "one variable, logistic regression effects model, sigma= 4 fit model:"
##
## Call:
## glm(formula = formulaL, family = binomial, data = trainData)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.0306 -1.1685 0.5217 1.1472 1.6129
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.45010 0.05285 8.516 <2e-16 ***
## x1 3.02567 0.20065 15.079 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2747.0 on 1999 degrees of freedom
## Residual deviance: 2447.3 on 1998 degrees of freedom
## AIC: 2451.3
##
## Number of Fisher Scoring iterations: 4
##
## [1] "one variable, logistic regression Noised 4 train mean deviance 1.76536175195365"
## [1] "one variable, logistic regression Noised 4 test mean deviance 1.75674472607196"
## [1] "effects model, jacknifed"
## [1] "one variable, logistic regression effects model, jackknifed fit model:"
##
## Call:
## glm(formula = formulaL, family = binomial, data = trainData)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.0811 -1.1892 0.4966 1.1600 1.5642
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.45308 0.05326 8.508 <2e-16 ***
## x1 2.99703 0.20478 14.636 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2747.0 on 1999 degrees of freedom
## Residual deviance: 2460.2 on 1998 degrees of freedom
## AIC: 2464.2
##
## Number of Fisher Scoring iterations: 4
##
## [1] "one variable, logistic regression jackknifed train mean deviance 1.77463669725858"
## [1] "one variable, logistic regression jackknifed test mean deviance 1.746225629925"
## [1] "********"
## [1] "one variable, logistic regression AverageManyNoisedModels"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.737 1.760 1.772 1.772 1.780 1.823
## [1] 0.0165008
## [1] "********"
## [1] "********"
## [1] "one variable, logistic regression JackknifeModel"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.737 1.761 1.772 1.772 1.779 1.819
## [1] 0.01604833
## [1] "********"
## [1] "********"
## [1] "one variable, logistic regression NaiveModel"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.737 1.760 1.772 1.772 1.780 1.823
## [1] 0.01688741
## [1] "********"
## [1] "********"
## [1] "one variable, logistic regression NoisedModel"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.744 1.761 1.774 1.776 1.785 1.895
## [1] 0.02226869
## [1] "********"
## [1] "********"
## [1] "one variable, logistic regression ObliviousModel"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.975 1.984 1.986 1.986 1.988 1.993
## [1] 0.003336783
## [1] "********"
## [1] "*************************************************************"
one variable plus noise variable, logistic regression
## [1] "*************************************************************"
## [1] "one variable plus noise variable, logistic regression"
## [1] "bSigmaBest 7"
## [1] "naive effects model"
## [1] "one variable plus noise variable, logistic regression naive effects model fit model:"
##
## Call:
## glm(formula = formulaL, family = binomial, data = trainData)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.5658 -0.9120 0.3055 0.8035 2.7112
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.68760 0.06161 11.16 <2e-16 ***
## x1 3.18452 0.23641 13.47 <2e-16 ***
## n1 2.45247 0.15572 15.75 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2747.0 on 1999 degrees of freedom
## Residual deviance: 1990.5 on 1997 degrees of freedom
## AIC: 1996.5
##
## Number of Fisher Scoring iterations: 6
##
## [1] "one variable plus noise variable, logistic regression naive effects model train mean deviance 1.43587337720022"
## [1] "one variable plus noise variable, logistic regression naive effects model test mean deviance 3.54303901440774"
## [1] "effects model, sigma= 7"
## [1] "one variable plus noise variable, logistic regression effects model, sigma= 7 fit model:"
##
## Call:
## glm(formula = formulaL, family = binomial, data = trainData)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.0177 -1.1986 0.5326 1.0979 1.6982
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.48151 0.05739 8.391 <2e-16 ***
## x1 3.03989 0.20336 14.948 <2e-16 ***
## n1 0.02137 0.01518 1.407 0.159
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2747.0 on 1999 degrees of freedom
## Residual deviance: 2454.7 on 1997 degrees of freedom
## AIC: 2460.7
##
## Number of Fisher Scoring iterations: 3
##
## [1] "one variable plus noise variable, logistic regression Noised 7 train mean deviance 1.77067562081881"
## [1] "one variable plus noise variable, logistic regression Noised 7 test mean deviance 1.78620816972643"
## [1] "effects model, jacknifed"
## [1] "one variable plus noise variable, logistic regression effects model, jackknifed fit model:"
##
## Call:
## glm(formula = formulaL, family = binomial, data = trainData)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.2012 -1.1757 0.5026 1.1657 1.5936
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.42346 0.05493 7.710 1.26e-14 ***
## x1 3.00699 0.20534 14.644 < 2e-16 ***
## n1 -0.05278 0.02435 -2.167 0.0302 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2747.0 on 1999 degrees of freedom
## Residual deviance: 2455.4 on 1997 degrees of freedom
## AIC: 2461.4
##
## Number of Fisher Scoring iterations: 4
##
## [1] "one variable plus noise variable, logistic regression jackknifed train mean deviance 1.77119992923416"
## [1] "one variable plus noise variable, logistic regression jackknifed test mean deviance 1.77521675815884"
## [1] "********"
## [1] "one variable plus noise variable, logistic regression AverageManyNoisedModels"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.744 1.760 1.773 1.773 1.782 1.820
## [1] 0.01470923
## [1] "********"
## [1] "********"
## [1] "one variable plus noise variable, logistic regression JackknifeModel"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.741 1.763 1.773 1.772 1.780 1.809
## [1] 0.01399957
## [1] "********"
## [1] "********"
## [1] "one variable plus noise variable, logistic regression NaiveModel"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 3.159 3.480 3.590 3.612 3.768 4.147
## [1] 0.2109597
## [1] "********"
## [1] "********"
## [1] "one variable plus noise variable, logistic regression NoisedModel"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.747 1.776 1.787 1.791 1.800 1.925
## [1] 0.02397365
## [1] "********"
## [1] "********"
## [1] "one variable plus noise variable, logistic regression ObliviousModel"
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.977 1.984 1.986 1.986 1.988 1.992
## [1] 0.003011528
## [1] "********"
## [1] "*************************************************************"